evergreen

For personal projects etc.

Evergreen skills for building Roblox games with AI coding agents (Claude Code, Codex, Cursor, Copilot): Studio MCP done safely, code on disk, Luau checks without Studio, Open Cloud, secure-code rules

Evergreen chart and graph skills: evidence-based chart knowledge base, chooser and builder CLI, five render targets

Agent Skill: lint, link and index AI-written markdown so it navigates in Obsidian and still renders on GitHub; vault setup; Obsidian CLI

MCP server aggregating application metadata (versions, download URLs, SHA256, silent switches, product codes) across winget, Chocolatey, Evergreen, and Silent Install HQ

Agent skill that makes prose read as written by a person, with a stdlib checker that proves it. Evergreen: re-researches the signs of AI writing on a schedule.

Evergreen AI-agent skills for making music: generate with local or hosted models, judge takes the agent cannot hear, master, loop for games, score to picture, and remember the listener's taste

Keep the repositories and packages you already own healthy: PR triage with cooldowns and lockfile checks, vulnerabilities, dependencies, modernization gaps, one digest. Agent skill + stdlib CLI. Evergreen.

Lint and de-conflict agent skill descriptions (Claude Code, Codex, Copilot, Cursor) and cut the skill catalog's token cost; evergreen

Evergreen 3D modelling skills for AI agents: real places from open data, image-to-3D, headless Blender, game-ready and printable models, all checked

Agent skill: write and update a repository's GitHub wiki, with every example verified against the published package (npm, NuGet). Evergreen.

An editor for your AI agent that marks the page and never writes on it: editorial letters, line, copy and proof suggestions, a provenance ledger of AI-written words, and only the author's choices applied.

Evergreen agent skills for modding MADNESS: Project Nexus: MadCards and mod folders, SceneTools levels and stages, ModTools assets and built-in Harmony code, with a stdlib helper that lints and deploys mods and reads the game's mod log

Evergreen skills that make AI agents answer history questions from sources, not memory: Wikidata full statements, footnotes followed to primary texts, dates computed in code, every citation checked, research packs for 3D and video

Cross-agent context health: hooks, a status line and a skill that warn when a session's context is big enough to degrade the model (Claude Code, Copilot CLI, VS Code, Codex, Gemini CLI, Cursor); evergreen per-model bands

Everlast: a knowledge layer for coding agents so nothing a session learned is lost across models, sessions and tools (ai-docs per project, a private vault for user-level facts, capture and resume skills)

Agent plugin: read public pages that refuse scripts (Reddit threads and more) through Gemini's URL context tool. Stdlib Python CLI, key and privacy setup skill. Never uses Google Search.

Name any subject and get an AI scout for it: verified communities and feeds, polite scanning, notes, signals, radar reports, and respectful questions to makers

Is this study legit? An evergreen AI-agent skill that appraises scientific papers and the news coverage around them, with a one-call Crossref/PubMed/Europe PMC metadata script.